EP1999578B1 - Representing loop branches in a branch history register with multiple bits - Google Patents

Representing loop branches in a branch history register with multiple bits Download PDF

Info

Publication number
EP1999578B1
EP1999578B1 EP07758843A EP07758843A EP1999578B1 EP 1999578 B1 EP1999578 B1 EP 1999578B1 EP 07758843 A EP07758843 A EP 07758843A EP 07758843 A EP07758843 A EP 07758843A EP 1999578 B1 EP1999578 B1 EP 1999578B1
Authority
EP
European Patent Office
Prior art keywords
loop
branch
bhr
branch instruction
iterations
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Not-in-force
Application number
EP07758843A
Other languages
German (de)
English (en)
French (fr)
Other versions
EP1999578A2 (en
Inventor
James Norris Dieffenderfer
Bohuslav c/o Qualcomm Incorporated RYCHLIK
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Qualcomm Inc
Original Assignee
Qualcomm Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Qualcomm Inc filed Critical Qualcomm Inc
Priority to EP12155320.0A priority Critical patent/EP2479662B1/en
Publication of EP1999578A2 publication Critical patent/EP1999578A2/en
Application granted granted Critical
Publication of EP1999578B1 publication Critical patent/EP1999578B1/en
Not-in-force legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/38—Concurrent instruction execution, e.g. pipeline or look ahead
    • G06F9/3836—Instruction issuing, e.g. dynamic instruction scheduling or out of order instruction execution
    • G06F9/3842—Speculative instruction execution
    • G06F9/3848—Speculative instruction execution using hybrid branch prediction, e.g. selection between prediction techniques
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30098—Register arrangements
    • G06F9/3012—Organisation of register space, e.g. banked or distributed register file
    • G06F9/3013—Organisation of register space, e.g. banked or distributed register file according to data content, e.g. floating-point registers, address registers
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/38—Concurrent instruction execution, e.g. pipeline or look ahead
    • G06F9/3802—Instruction prefetching
    • G06F9/3804—Instruction prefetching for branches, e.g. hedging, branch folding
    • G06F9/3806—Instruction prefetching for branches, e.g. hedging, branch folding using address prediction, e.g. return stack, branch history buffer

Definitions

  • the present disclosure relates generally to the field of processors and in particular to a method of representing loop branches in a branch history register with multiple bits.
  • Microprocessors perform computational tasks in a wide variety of applications. Improved processor performance is almost always desirable, to allow for faster operation and/or increased functionality through software changes. In common embedded applications, such as portable electronic devices, conserving power is also desirable.
  • processors employ a pipelined architecture, where sequential instructions, each having multiple execution steps, are overlapped in execution. For maximum performance, the instructions should flow continuously through the pipeline. Any situation that causes instructions to stall in the pipeline detrimentally affects performance. If instructions must be flushed from the pipeline and subsequently re-fetched, both performance and power consumption may suffer.
  • conditional branch instructions the actual branching behavior of which is commonly not known until the instruction is evaluated deep in the pipeline.
  • branch prediction To avoid a stall that may result from waiting for actual evaluation of the branch instruction, common modern processors employ some form of branch prediction, whereby the branching behavior of conditional branch instructions is predicted early in the pipeline. Based on the predicted branch evaluation, the processor speculatively fetches (prefetches) and executes instructions from a predicted address - either the branch target address (if the branch is predicted taken) or the next sequential address after the branch instruction (if the branch is predicted not taken).
  • branch prediction techniques include both static and dynamic predictions.
  • the likely behavior of some branch instructions can be statically predicted by a programmer and/or compiler.
  • One example is an error checking routine. Common code executes properly, and errors are rare. Hence, the branch instruction implementing a "branch on error” function will evaluate "not taken" a very high percentage of the time.
  • Such an instruction may include a static branch prediction bit in the op code, set by a programmer or compiler with knowledge of the common likely outcome of the branch condition.
  • Dynamic prediction is generally based on the branch evaluation history (and in some cases the branch prediction accuracy history) of the branch instruction being predicted and/or other branch instructions in the same code. Extensive analysis of actual code indicates that recent past branch evaluation patterns may be a good indicator of the evaluation of future branch instructions.
  • BHR Branch History Register
  • the BHR 30 comprises a shift register.
  • the common recent branch evaluation result is shifted in (for example, a 1 indicating taken and a 0 indicating not taken), with the oldest past evaluation in the register being displaced.
  • a processor may maintain a local BHR 100 for each branch instruction.
  • a BHR 100 may contain the recent past evaluations of all conditional branch instructions, sometimes known in the art as a global BHR, or GHR.
  • BHR refers to both local and global Branch History Registers.
  • the BHR 100 may index a Branch Predictor Table (BPT) 102, which again may be local or global.
  • BPT Branch Predictor Table
  • the BHR 100 may index the BPT 102 directly, or may be combined with other information, such as the Program Counter (PC) of the branch instruction in BPT index logic 104.
  • PC Program Counter
  • Other inputs to the BPT index logic 104 may additionally be utilized.
  • the BPT index logic 104 may concatenate the inputs (known in the art as gselect), XOR the inputs (gshare), perform a hash function, or combine or transform the inputs in a variety of ways.
  • the BPT 102 may comprise a plurality of saturation counters, the MSBs of which serve as bimodal branch predictors.
  • each table entry may comprise a 2-bit counter that assumes one of four states, each assigned a weighted prediction value, such as:
  • the counter increments each time a corresponding branch instruction evaluates "taken” and decrements each time the instruction evaluates "not taken.”
  • the MSB of the counter is a bimodal branch predictor; it will predict a branch to be either taken or not taken, regardless of the strength or weight of the underlying prediction.
  • a saturation counter reduces the prediction error of an infrequent branch evaluation direction. A branch that consistently evaluates one way will saturate the counter. An infrequent evaluation the other way will alter the counter value (and the strength of the prediction), but not the bimodal prediction value. Thus, an infrequent evaluation will only mispredict once, not twice.
  • the table of saturation counters is an illustrative example only; in general, a BHT may index a table containing a variety of branch prediction mechanisms.
  • the BHR 100 either alone or in combination with other information such as the branch instruction PC - indexes the BPT 102 to obtain branch predictions.
  • the branch instruction being predicted is correlated to past branch behavior - its own past behavior in the case of a local BHR 100 and the behavior of other branch instructions in the case of a global BHR 100. This correlation is the key to accurate branch predictions, at least in the case of highly repetitive code.
  • Figure 1 depicts branch evaluations being stored in the BHR 100 - that is, the actual evaluation of a conditional branch instruction, which may only be known deep in the pipeline, such as in an execute pipe stage. While this is the ultimate result, in practice, common high performance processors store the predicted branch evaluation from the BPT 102 in the BHR 100, and correct the BHR 100 later as part of a misprediction recovery operation if the prediction turns out to be erroneous.
  • the drawing figures do not reflect this implementation feature, for clarity.
  • a common code structure that may reduce the efficacy a branch predictor employing a BHR 100 is the loop.
  • a loop ends with a conditional branch instruction that tests a loop-ending condition, such as whether an index variable that is incremented each time through the loop has reached a loop-ending value. If not, execution branches back to the beginning of the loop for another iteration, and another loop-ending conditional branch evaluation.
  • the "taken" backwards branches of the loop-ending branch instruction partially or fully saturate the BHR 100.
  • the number of loop iterations equals or exceeds the BHR 100 width
  • an n -bit BHR will contain precisely n -1 ones (taken) followed by a single zero (not taken), corresponding to a long series of branch taken evaluations resulting from the loop iterations, and ending with a single not-taken branch evaluation when the loop terminates. This effectively destroys the efficacy of the BHR 100, as all correlations with prior branch evaluations (for either a local or global BHR 100) are lost.
  • the BHR 100 will likely map to the same BPT 102 entry for a given branch instruction (depending on the other inputs to the BPT index logic 104), rather than to an entry containing a branch prediction that reflects the correlation of the branch instruction to prior branch evaluations.
  • the saturated BHR 100 may increase aliasing in the BPT 102. That is, all branch instructions following loops with common iterations will map to the same BPT 102 entry, if the BHR 100 directly indexes the BPT 102. Even where the BHR 100 is combined with other information, the chance of aliasing is increased. This adversely impacts prediction accuracy not only for the branch instruction following the loop, but also for all of the branch instructions that alias to its entry in the BPT 102.
  • the branch instruction will map to a much larger number of entries in the BPT 102 to capture the same correlation with prior branch evaluations, requiring a larger BPT 102 to support the same accuracy for the same number of branch instructions than would be required without the loop-ending branch affecting the BHR 30.
  • the branch predictors in the BPT 102 will take longer to "train,” increasing the amount of code that must execute before the BPT 102 begins to provide accurate branch predictions.
  • Patent application Serial No. 11/066,508 assigned to the assignee of the present application and incorporated herein by reference, proposes to suppress updating the BHR 100 for loop-ending branch instructions. This resolves common of the deleterious effects of BHR 100 saturation or partial saturation on branch prediction accuracy. However, it fails to capture and exploit correlations that may exist between loop behavior and subsequent branch evaluation.
  • the evaluation of a branch instruction may be correlated to the number of iterations of a preceding loop.
  • a scientific program may capture data points in a loop and, following the loop, branch to a statistical analysis subroutine where the captured data is analyzed.
  • the loop iterates relatively few times, capturing few data points and yielding a small sample, statistical analysis may be unreliable, and may be skipped.
  • the evaluation of the conditional instruction branching to the statistical analysis subroutine is strongly correlated to the number of iterations of the data acquisition loop.
  • an application may use a loop structure to search through a list, transaction log, history file, or similar data structure. If an item matching the search parameters appears frequently in the list, relatively few loop iterations will be required to locate the item. Consequently, an item appearing infrequently may require a large number of iterations through the search loop.
  • the evaluation of a subsequent branch instruction may be correlated to the frequency with which a particular item appears in the list, and hence correlated to the number of loop iterations required to locate the item.
  • a count of the number of iterations of the loop is maintained, and a multi-bit value indicative of the loop iteration count is stored in the BHR.
  • the multi-bit value may comprise the actual loop count, in which case the number of bits is variable.
  • the number of bits is fixed (e.g., two) and loop iteration counts are mapped to a multi-bit value by comparison to thresholds. Separate iteration counts may be maintained for nested loops, and a multi-bit value stored in the BHR may indicate a loop iteration count of only an inner loop, only the outer loop, or both.
  • One embodiment relates to a method of branch prediction.
  • a property of a branch instruction associated with a loop is identified.
  • a multi-bit value is stored in a BHR, the multi-bit value indicative of the number of iterations of a loop associated with the branch instruction.
  • the processor includes a branch predictor operative to predict the evaluation of conditional branch instructions, and an instruction execution pipeline operative to speculatively fetch and execute instructions based on a prediction from the branch predictor.
  • the processor also includes a BHR operative to store the evaluation of conditional branch instructions, and a loop counter operative to count the number of iterations of a code loop.
  • the processor further includes a control circuit operative to store in the BHR a multi-bit value indicative of the number of iterations of a loop associated with a conditional branch instruction.
  • Figure 1 is a functional block diagram of a prior art branch predictor circuit.
  • Figure 2 is a functional block diagram of a processor.
  • Figure 3 is a flow diagram of a method of executing a branch instruction.
  • Figure 4 is a functional block diagram of a branch predictor circuit including one or more Last Branch PC registers and Loop Counters.
  • Figure 2 depicts a functional block diagram of a processor 10.
  • the processor 10 executes instructions in an instruction execution pipeline 12 according to control logic 14.
  • the pipeline 12 may be a superscalar design, with multiple pipelines.
  • the pipeline 12 includes various registers or latches 16, organized in pipe stages, and one or more Arithmetic Logic Units (ALU) 18.
  • ALU Arithmetic Logic Units
  • a General Purpose Register (GPR) file 20 provides registers comprising the top of the memory hierarchy.
  • GPR General Purpose Register
  • the pipeline 12 fetches instructions from an instruction cache (I-cache) 22, with memory address translation and permissions managed by an Instruction-side Translation Lookaside Buffer (ITLB) 24.
  • I-cache instruction cache
  • ITLB Instruction-side Translation Lookaside Buffer
  • a branch predictor 26 predicts the branch behavior, and provides the prediction to an instruction prefetch unit 28.
  • the instruction prefetch unit 28 speculatively fetches instructions from the 1-cache 22, at a branch target address calculated in the pipeline 12 for "taken" branch predictions, or at the next sequential address for branches predicted “not taken.” In either case, the prefetched instructions are loaded into the pipeline 12 for speculative execution.
  • the branch predictor 26 includes a Branch History Register (BHR) 30, a Branch Predictor Table (BPT) 32, BPT index logic 34, and BHR update logic 36.
  • the branch predictor 26 may additionally include one or more Last Branch PC registers 38, and one or more loop counters 37 providing inputs to the BHR update logic 36.
  • Data is accessed from a data cache (D-cache) 40, with memory address translation and permissions managed by a main Translation Lookaside Buffer (TLB) 42.
  • the TLB 42 may be a dedicated data TLB, or may comprise an integrated TLB that manages address translations and permissions for both instructions and data.
  • the I-cache 22 and D-cache 40 may be integrated, or unified. Misses in the I-cache 22 and/or the D-cache 40 cause an access to main (off-chip) memory 44, under the control of a memory interface 46.
  • the processor 10 may include an Input/Output (I/O) interface 48, controlling access to various peripheral devices 50, 52.
  • I/O Input/Output
  • the processor 10 may include a second-level (L2) cache for either or both the I and D caches 22, 40.
  • L2 second-level cache for either or both the I and D caches 22, 40.
  • one or more of the functional blocks depicted in the processor 10 may be omitted from a particular embodiment.
  • branch prediction accuracy is improved by correlating branch evaluation to both previous branch evaluations and the iteration count of previous loops. This is accomplished by storing a multi-bit value indicative of loop iteration count(s) in the BHR 30. By mapping potentially large loop iteration counts to relatively few bits, characteristic information about the loop is preserved while preventing the loop-ending branches from fully or partially saturating one or more BHRs 30 in the branch predictor 26.
  • An instruction is fetched and decoded (block 52). If the instruction is a not conditional branch instruction (block 53), it proceeds in the pipeline and the process returns to fetch the next instruction (block 52). If the instruction is a conditional branch instruction (block 53), a determination is made whether the branch is a loop-ending branch (block 54). If not, when the branch evaluates (such as in an execute stage of the pipeline) the BHR 30 is updated to record the branch evaluation using a single bit (block 56), i.e., whether the branch instruction evaluated as "taken" or "not taken.” Execution then continues (block 64) at the branch target address or the next sequential address, respectively.
  • a loop iteration count is incremented (block 58).
  • a multi-bit value indicative of the number of loop iterations is written to the BHR 30 (block 62).
  • the loop iteration counter is then cleared (block 63) and execution continues (block 64) at the next sequential address.
  • Both the query (block 54) - that is, identifying a branch instruction as a loop-ending branch instruction - and maintaining the loop iteration count (block 58) may be accomplished in a variety of ways.
  • one or more Loop Counters (LC) 37 are incremented when a conditional branch instruction that is determined to be a loop-ending branch evaluates "taken.”
  • the value of the LC 37, or a value derived therefrom (as discussed more fully herein) is written to the BHR 30 when the loop-ending branch evaluates "not taken," indicating an exit of the loop.
  • the LC 37 is also cleared at that time.
  • Loops iterate by branching backwards from the end of the loop to the beginning of the loop.
  • every conditional branch instruction with a branch target address less than the branch instruction address, or PC - that is, a backwards branch - is assumed to be a loop-ending branch instruction.
  • This method of identifying Loop-Ending branches offers simplicity.
  • the branch instruction PC is compared to the branch target address (BTA) in Loop-Ending Branch Detection logic 35 when the branch instruction is actually evaluated in the pipeline. If BTA ⁇ PC, a Loop Counter (LC) 37 is incremented.
  • BTA branch target address
  • LC Loop Counter
  • a Last Branch PC (LBPC) register 38 stores the PC of the last branch instruction executed.
  • LBPC Last Branch PC
  • Code that includes conditional branch instructions within a loop can complicate loop-ending branch detection. For example, common algorithms for searching an array or linked list create a loop. A branch within the loop checks for a match and exits the loop early if a match is found, while a loop-ending branch checks for the end of the array or linked list. In this case, a single LBPC 38 cannot detect the loop-ending branch, because it will be alternately overwritten by the two branch PCs. Accordingly, in one embodiment, two or more LBPC registers 38 may be provided (as depicted in Fig. 4 ), with the PCs of successively evaluated branch instructions stored in corresponding LBPC registers (LBPC 0 , LBPC 1 , ... LBPC M ) 38. This allows for the detection of loop-ending branches even in the presence of other conditional branches within the loop.
  • Loop-ending branch instructions may also be statically marked by a compiler or assembler.
  • a compiler generates a particular type of branch instruction that is only used for loop-ending branches, for example, "BRLP".
  • the BRLP instruction is recognized, and a LC 37 is incremented every time a BRLP instruction evaluates as taken in an execution pipe stage.
  • a compiler or assembler may embed a loop-ending branch indication in a branch instruction, such as by setting one or more predefined bits in the instruction op code.
  • the loop-ending branch bits are detected, and a LC 37 is incremented when that branch instruction evaluates as taken in an execute pipe stage.
  • a Loop-Ending Indicator may be provided to the Loop-Ending Branch Detection logic 35. Static identification of loop-ending branches reduces hardware and computational complexity by moving the loop-ending identification function into the compiler or assembler.
  • an indication of the number of times the loop iterates is stored in the BHR 30, to exploit any correlation between the number of loop iterations and subsequent conditional branch instruction evaluation.
  • the actual count of loop iterations is stored in the BHR 30. That is, the multi-bit count from, e.g ., a LC 37 is shifted into the BHR 30, displacing prior branch evaluations in the MSBs of the BHR 30. For very large loops, this count will require log 2 (n) bits for n loop iterations, displacing a corresponding number of prior branch evaluations. However, even for large loop iteration counts, the binary representation is a significant compression of the loop-ending branch behavior stored in prior art branch history registers ( i.e., n-1 ones followed by a single zero). In one embodiment, the iteration count is accumulated in a LC 37 and shifted into the BHR 30.
  • At least the lower k bits of the BHR 30 may be implemented as a dual-mode shift-register/counter, and the loop iteration count may be incremented directly in the BHR 30 (with prior BHR 30 contents shifting left as the count grows).
  • a fixed, multi-bit indication of the number of loop iterations is stored in the in the BHR 30.
  • two bits may be allocated to record an indication of the number of loop iterations, with the following meanings:
  • threshold values - between small and moderate (00 - > 01); between moderate and large (01 -> 10), and between large and very large (10 -> 11) - are used to map the actual loop iteration count to a two-bit representation, as indicated by Thresholding logic 39.
  • the threshold values may be fixed for a given processor implementation, such as for certain embedded applications where common loop sizes are generally known.
  • the mapping may be scalable, with the threshold values read from GPRs 20, a table in memory 44, or the like.
  • a fixed multi-bit indication of the number of loop iterations may provide sufficient indication of the loop iteration count to allow for meaningful correlation with branch evaluation behavior, while at the same time limiting the number of prior branch evaluations displaced from the BHR 30 by loop-ending branches.
  • actual loop iteration counts may be mapped to any fixed number of bits, with the corresponding number of threshold values, e.g ., three bits and seven thresholds; four bits and fifteen thresholds; etc., to achieve the desired or required balance between loop iteration count accuracy and branch evaluation displacements in the BHR 30.
  • a loop may contain one or more nested, or inner, loops. Accordingly, in one embodiment, a plurality of LC counters (LC 0 , LC 1 , ... LC M ) 37 are provided, which may correspond to the number of LBPC 38 registers. A LC 37 is incremented upon a taken evaluation and a match with the relevant LBPC M 38, to maintain the iteration counts of nested loops.
  • LC counters LC 0 , LC 1 , ... LC M
  • the BHR 30 may become partially or completely saturated by the time the outer loop iterates relatively few times. Accordingly, in one embodiment, once a loop is determined to be an inner loop, subsequent iteration counts after an initial execution of the inner loop may be suppressed from updating the BHR 30. In this embodiment, by the time the outer loop exits, the BHR 30 will include an indication of the iteration count of the inner loop only for the first iteration of the outer loop, and an indication of the iteration count of the outer loop.
  • the inner loop iteration count may be discarded by overwriting its value in the BHR 30 with an indication of the outer loop iteration count. This preserves the greatest number of prior branch evaluations in the BHR 30, while allowing for correlation of subsequent branch evaluations with the iteration count of the outer loop.
  • the outer loop iteration count may be discarded by suppressing an update of the BHR 30 when the outer loop exits. This again preserves the greatest number of prior branch evaluations in the BHR 30, while allowing for correlation of subsequent branch evaluations with the iteration count of the inner loop, which may be more predictive of subsequent branch evaluation in some applications.
  • a conditional branch instruction has common properties, including for example the branch instruction address or PC, the instruction type, and the presence, vel non, of indicator bits in the op code.
  • properties of the branch operation, and/or properties of the program that relate to the branch are considered properties of the branch instruction. For example, whether the branch instruction PC matches the contents of one or more LBPC registers 38, and whether the branch target address is forward or backward relative to the branch instruction PC, are properties of the branch instruction.

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Advance Control (AREA)
  • Executing Machine-Instructions (AREA)
EP07758843A 2006-03-17 2007-03-19 Representing loop branches in a branch history register with multiple bits Not-in-force EP1999578B1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP12155320.0A EP2479662B1 (en) 2006-03-17 2007-03-19 Representing loop branches in a branch history register with multiple bits

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US11/378,712 US8904155B2 (en) 2006-03-17 2006-03-17 Representing loop branches in a branch history register with multiple bits
PCT/US2007/064331 WO2007109631A2 (en) 2006-03-17 2007-03-19 Branch history register for loop branches

Related Child Applications (2)

Application Number Title Priority Date Filing Date
EP12155320.0A Division EP2479662B1 (en) 2006-03-17 2007-03-19 Representing loop branches in a branch history register with multiple bits
EP12155320.0 Division-Into 2012-02-14

Publications (2)

Publication Number Publication Date
EP1999578A2 EP1999578A2 (en) 2008-12-10
EP1999578B1 true EP1999578B1 (en) 2012-05-16

Family

ID=38175561

Family Applications (2)

Application Number Title Priority Date Filing Date
EP07758843A Not-in-force EP1999578B1 (en) 2006-03-17 2007-03-19 Representing loop branches in a branch history register with multiple bits
EP12155320.0A Not-in-force EP2479662B1 (en) 2006-03-17 2007-03-19 Representing loop branches in a branch history register with multiple bits

Family Applications After (1)

Application Number Title Priority Date Filing Date
EP12155320.0A Not-in-force EP2479662B1 (en) 2006-03-17 2007-03-19 Representing loop branches in a branch history register with multiple bits

Country Status (11)

Country Link
US (1) US8904155B2 (enrdf_load_stackoverflow)
EP (2) EP1999578B1 (enrdf_load_stackoverflow)
JP (1) JP5384323B2 (enrdf_load_stackoverflow)
KR (1) KR101031938B1 (enrdf_load_stackoverflow)
CN (2) CN106997286B (enrdf_load_stackoverflow)
BR (1) BRPI0709558A2 (enrdf_load_stackoverflow)
CA (1) CA2643969A1 (enrdf_load_stackoverflow)
ES (1) ES2388309T3 (enrdf_load_stackoverflow)
MX (1) MX2008011752A (enrdf_load_stackoverflow)
RU (1) RU2447486C2 (enrdf_load_stackoverflow)
WO (1) WO2007109631A2 (enrdf_load_stackoverflow)

Families Citing this family (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7707398B2 (en) * 2007-11-13 2010-04-27 Applied Micro Circuits Corporation System and method for speculative global history prediction updating
US9110683B2 (en) * 2008-08-15 2015-08-18 Apple Inc. Predicting branches for vector partitioning loops when processing vector instructions
US8433885B2 (en) * 2009-09-09 2013-04-30 Board Of Regents Of The University Of Texas System Method, system and computer-accessible medium for providing a distributed predicate prediction
US8073556B2 (en) * 2009-12-16 2011-12-06 General Electric Company System and method for controlling a machine
US9176737B2 (en) * 2011-02-07 2015-11-03 Arm Limited Controlling the execution of adjacent instructions that are dependent upon a same data condition
CN103999045B (zh) * 2011-12-15 2017-05-17 英特尔公司 使用混洗表和混合表经由矢量指令优化程序循环的方法
US9304776B2 (en) * 2012-01-31 2016-04-05 Oracle International Corporation System and method for mitigating the impact of branch misprediction when exiting spin loops
US9389860B2 (en) * 2012-04-02 2016-07-12 Apple Inc. Prediction optimizations for Macroscalar vector partitioning loops
US20150074353A1 (en) * 2013-09-06 2015-03-12 Futurewei Technologies, Inc. System and Method for an Asynchronous Processor with Multiple Threading
US9639368B2 (en) * 2014-06-13 2017-05-02 International Business Machines Corporation Branch prediction based on correlating events
GB2540221B (en) * 2015-07-08 2020-01-01 Mips Tech Llc Check pointing a shift register
US9875104B2 (en) 2016-02-03 2018-01-23 Google Llc Accessing data in multi-dimensional tensors
CN108062236A (zh) * 2016-11-07 2018-05-22 杭州华为数字技术有限公司 一种软硬件协同分支指令预测方法及装置
CN106528049B (zh) * 2016-11-17 2019-06-11 上海兆芯集成电路有限公司 在多存储体条件分支预测器中用于更新事件的随机数产生
CN108255518B (zh) * 2016-12-29 2020-08-11 展讯通信（上海）有限公司 处理器及循环程序分支预测方法
US10248908B2 (en) * 2017-06-19 2019-04-02 Google Llc Alternative loop limits for accessing data in multi-dimensional tensors
GB2568776B (en) 2017-08-11 2020-10-28 Google Llc Neural network accelerator with parameters resident on chip
CN108595210B (zh) 2018-04-09 2021-12-10 杭州中天微系统有限公司 实现零开销循环的处理器
US11216279B2 (en) * 2018-11-26 2022-01-04 Advanced Micro Devices, Inc. Loop exit predictor
US11294681B2 (en) * 2019-05-31 2022-04-05 Texas Instruments Incorporated Processing device with a microbranch target buffer for branch prediction using loop iteration count
US12112173B2 (en) * 2020-12-21 2024-10-08 Intel Corporation Context-based loop branch prediction
CN116737240B (zh) 2022-03-02 2024-08-06 腾讯科技（深圳）有限公司 分支预测方法、装置、处理器、介质及设备
CN115113934B (zh) * 2022-08-31 2022-11-11 腾讯科技（深圳）有限公司 指令处理方法、装置、程序产品、计算机设备和介质
US12217059B1 (en) 2023-03-30 2025-02-04 Advanced Micro Devices, Inc. Systems and methods for programmed branch predictors

Family Cites Families (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS635442A (ja) * 1986-06-26 1988-01-11 Matsushita Electric Ind Co Ltd プログラムル−プ検出記憶装置
JPH0635442A (ja) 1992-07-13 1994-02-10 Fuji Xerox Co Ltd データ変換装置
JP3114833B2 (ja) 1993-04-20 2000-12-04 本田技研工業株式会社 油圧無段変速機付き車両の変速操作装置
US5909573A (en) * 1996-03-28 1999-06-01 Intel Corporation Method of branch prediction using loop counters
US5875325A (en) * 1996-09-19 1999-02-23 International Business Machines Corporation Processor having reduced branch history table size through global branch history compression and method of branch prediction utilizing compressed global branch history
JP3760041B2 (ja) * 1996-12-09 2006-03-29 松下電器産業株式会社 分岐予測する情報処理装置
GB2323190B (en) * 1997-03-14 2001-09-19 Nokia Mobile Phones Ltd Executing nested loops
US6374349B2 (en) * 1998-03-19 2002-04-16 Mcfarling Scott Branch predictor with serially connected predictor stages for improving branch prediction accuracy
US6223280B1 (en) * 1998-07-16 2001-04-24 Advanced Micro Devices, Inc. Method and circuit for preloading prediction circuits in microprocessors
RU2236700C2 (ru) 2000-03-02 2004-09-20 Открытое акционерное общество "Московская телекоммуникационная корпорация" Способ анализа и прогнозирования развития динамической системы и ее отдельных элементов
US6766443B2 (en) * 2001-05-17 2004-07-20 International Business Machines Corporation Compression of execution path history to improve branch prediction accuracy
TW503997U (en) 2001-12-14 2002-09-21 Hon Hai Prec Ind Co Ltd Optical switch device
US7439982B2 (en) * 2002-05-31 2008-10-21 Envivio, Inc. Optimized scene graph change-based mixed media rendering
JP4243463B2 (ja) 2002-08-19 2009-03-25 株式会社半導体理工学研究センター 命令スケジューリングのシミュレーション方法とシミュレーションシステム
US20040123075A1 (en) * 2002-12-19 2004-06-24 Yoav Almog Extended loop prediction techniques
US20050283593A1 (en) * 2004-06-18 2005-12-22 Vladimir Vasekin Loop end prediction
JP4362096B2 (ja) 2004-09-30 2009-11-11 富士通株式会社 情報処理装置，リプレース方法，リプレースプログラム及びリプレースプログラムを記録したコンピュータ読取可能な記録媒体
US20060190710A1 (en) * 2005-02-24 2006-08-24 Bohuslav Rychlik Suppressing update of a branch history register by loop-ending branches
US20070079288A1 (en) * 2005-09-30 2007-04-05 Chad Willwerth System and method for capturing filtered execution history of executable program code

Also Published As

Publication number Publication date
BRPI0709558A2 (pt) 2011-07-19
KR20080097242A (ko) 2008-11-04
RU2008141155A (ru) 2010-04-27
CN106997286A (zh) 2017-08-01
ES2388309T3 (es) 2012-10-11
JP2009530754A (ja) 2009-08-27
EP2479662A1 (en) 2012-07-25
CN101401065A (zh) 2009-04-01
CA2643969A1 (en) 2007-09-27
CN101401065B (zh) 2016-11-02
US20070220239A1 (en) 2007-09-20
MX2008011752A (es) 2008-09-26
RU2447486C2 (ru) 2012-04-10
KR101031938B1 (ko) 2011-04-29
WO2007109631A3 (en) 2007-11-29
EP2479662B1 (en) 2016-01-06
WO2007109631A2 (en) 2007-09-27
CN106997286B (zh) 2019-08-13
JP5384323B2 (ja) 2014-01-08
EP1999578A2 (en) 2008-12-10
US8904155B2 (en) 2014-12-02

Similar Documents

Publication Publication Date Title
EP1999578B1 (en) Representing loop branches in a branch history register with multiple bits
KR100930199B1 (ko) 루프―종료 브랜치들에 의한 브랜치 히스토리 레지스터의업데이트 억제
JP2011100466A5 (enrdf_load_stackoverflow)
JP2008532142A5 (enrdf_load_stackoverflow)
JP2009530754A5 (enrdf_load_stackoverflow)
US20020144101A1 (en) Caching DAG traces
KR20070118135A (ko) 인덱스당 2개 이상의 분기 타겟 어드레스를 저장하는 분기타겟 어드레스 캐시
US7984279B2 (en) System and method for using a working global history register
US7827392B2 (en) Sliding-window, block-based branch target address cache
JP2009536770A (ja) ブロックに基づく分岐先アドレスキャッシュ
KR101048258B1 (ko) 가변 길이 명령 세트의 브랜치 명령의 최종 입도와 캐싱된 브랜치 정보의 관련
US8086831B2 (en) Indexed table circuit having reduced aliasing
HK1112984A (en) Suppressing update of a branch history register by loop-ending branches
HK1112086A (en) Branch target address cache storing two or more branch target addresses per index

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

17P Request for examination filed

Effective date: 20080916

AK Designated contracting states

Kind code of ref document: A2

Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LI LT LU LV MC MT NL PL PT RO SE SI SK TR

RIN1 Information on inventor provided before grant (corrected)

Inventor name: DIEFFENDERFER, JAMES NORRIS

Inventor name: RYCHLIK, BOHUSLAV C/O QUALCOMM INCORPORATED

17Q First examination report despatched

Effective date: 20100205

GRAP Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOSNIGR1

GRAS Grant fee paid

Free format text: ORIGINAL CODE: EPIDOSNIGR3

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

DAX Request for extension of the european patent (deleted)
AK Designated contracting states

Kind code of ref document: B1

Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LI LT LU LV MC MT NL PL PT RO SE SI SK TR

REG Reference to a national code

Ref country code: GB

Ref legal event code: FG4D

REG Reference to a national code

Ref country code: CH

Ref legal event code: EP

REG Reference to a national code

Ref country code: AT

Ref legal event code: REF

Ref document number: 558372

Country of ref document: AT

Kind code of ref document: T

Effective date: 20120615

REG Reference to a national code

Ref country code: IE

Ref legal event code: FG4D

REG Reference to a national code

Ref country code: DE

Ref legal event code: R096

Ref document number: 602007022703

Country of ref document: DE

Effective date: 20120712

REG Reference to a national code

Ref country code: NL

Ref legal event code: VDEP

Effective date: 20120516

REG Reference to a national code

Ref country code: ES

Ref legal event code: FG2A

Ref document number: 2388309

Country of ref document: ES

Kind code of ref document: T3

Effective date: 20121011

REG Reference to a national code

Ref country code: LT

Ref legal event code: MG4D

Effective date: 20120516

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: LT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120516

Ref country code: IS

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120916

Ref country code: SE

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120516

Ref country code: FI

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120516

Ref country code: CY

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120516

Ref country code: PL

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120516

REG Reference to a national code

Ref country code: AT

Ref legal event code: MK05

Ref document number: 558372

Country of ref document: AT

Kind code of ref document: T

Effective date: 20120516

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: GR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120817

Ref country code: PT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120917

Ref country code: SI

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120516

Ref country code: LV

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120516

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: BE

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120516

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: CZ

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120516

Ref country code: NL

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120516

Ref country code: RO

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120516

Ref country code: AT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120516

Ref country code: EE

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120516

Ref country code: SK

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120516

Ref country code: DK

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120516

PLBE No opposition filed within time limit

Free format text: ORIGINAL CODE: 0009261

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT

26N No opposition filed

Effective date: 20130219

REG Reference to a national code

Ref country code: DE

Ref legal event code: R097

Ref document number: 602007022703

Country of ref document: DE

Effective date: 20130219

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: BG

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120816

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: MC

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20130331

REG Reference to a national code

Ref country code: CH

Ref legal event code: PL

REG Reference to a national code

Ref country code: IE

Ref legal event code: MM4A

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: LI

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20130331

Ref country code: IE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20130319

Ref country code: CH

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20130331

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: MT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120516

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: TR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120516

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: HU

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT; INVALID AB INITIO

Effective date: 20070319

Ref country code: LU

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20130319

REG Reference to a national code

Ref country code: FR

Ref legal event code: PLFP

Year of fee payment: 10

REG Reference to a national code

Ref country code: FR

Ref legal event code: PLFP

Year of fee payment: 11

REG Reference to a national code

Ref country code: FR

Ref legal event code: PLFP

Year of fee payment: 12

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: GB

Payment date: 20180223

Year of fee payment: 12

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: FR

Payment date: 20180223

Year of fee payment: 12

Ref country code: IT

Payment date: 20180315

Year of fee payment: 12

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: ES

Payment date: 20180403

Year of fee payment: 12

GBPC Gb: european patent ceased through non-payment of renewal fee

Effective date: 20190319

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: GB

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20190319

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: IT

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20190319

Ref country code: FR

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20190331

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: DE

Payment date: 20200214

Year of fee payment: 14

REG Reference to a national code

Ref country code: ES

Ref legal event code: FD2A

Effective date: 20200727

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: ES

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20190320

REG Reference to a national code

Ref country code: DE

Ref legal event code: R119

Ref document number: 602007022703

Country of ref document: DE

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: DE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20211001